An Investigation of the Effects of Feeding Inhibitors on Colorado Potato Beetle, Leptinotarsa decemlineata (Say), and Fall Armyworm, Spodoptera frugiperda (J.E. Smith) Larvae

Limonoids constitute a potentially potent and abundant class of naturally-occurring insect antifeedants. The citrus limonoids are available in large quantities as waste products from the citrus industry and are easily isolated from the fruits\u27 seeds. In this Thesis, two insect species were used to investigate antifeedant activity: the Colorado potato beetle, Leptinotarsa decemlineata (Say) (Coleoptera: Chrysomelidae) and the fall armyworm, Spodoptera frugiperda (J. E. Smith) (Leptinotarsa: Noctuidae). Limonin, the principal bitter component of citrus, was found to have two different modes of action: limonin had a post-ingestive effect on Colorado potato beetle larvae, whereas limonin acted as a feeding deterrent via the peripheral nervous system of fall armyworm larvae. Attempts to determine the specific chemosensory organs responsible for the observed deterrency in fall armyworm larvae were unsuccessful. The results of structure-activity studies with citrus limonoids indicated that both the furan system of the D-ring and the epoxide on the C-ring were necessary for antifeedant activity against Colorado potato beetle and fall armyworm larvae. A model incorporating the C and D-rings and furan system was as active as limonin against Colorado potato beetle and fall armyworm larvae. The stereochemistry of the model\u27s functional groups was not important in determining antifeedant activity. For the A, A$\sp\prime$ and B-rings of limonoids, however, structure-activity relationships differed for the two insects. An $\propto,\beta$-unsaturated lactone A-ring significantly increased antifeedant activity against the fall armyworm, but not against the Colorado potato beetle. The addition of a hydroxyl group at the 7 position of the B-ring increased activity against Colorado potato beetle larvae, but not against fall armyworm larvae. Limonin was tested at 32 ug/disk at various time periods throughout the fifth stadium of the fall armyworm in both choice and no-choice situations. The effectiveness of limonin\u27s protection in choice assays was high and remained constant for the length of the experiment. During the early stages of the experiment, larvae did not show a use-dependent reduction in response to limonin in no-choice assays. Late in the experiment, however, the amount of limonin-treated material consumed in no-choice assays increased